Large discrepancies in otomycosis treatment in private ear, nose, and throat clinics in Denmark

Otomycosis is a fungal infection of the external ear canal that may involve the middle ear in case of tympanic membrane perforation and also may involve the auricle [1-3]. The main causative fungi are yeasts (Candida spp.), molds (Aspergillus spp.) and dermatophytes [1, 4, 5]. Fungi cause 7-15% of external otitis and the treatment is often long and cumbersome [6-8]. Predisposing factors include a warm humid climate, frequent swimming, eczema, excessive use of cotton tips, a narrow ear canal, allergy, chronic drainage, irradiation, obstructing ear wax, a radical cavity after mastoidectomy, dermatomycosis, a weakened immune system, earplugs and secondary to prolonged use of topical antibacterial treatment [1, 6, 9, 10]. Immunocompromised patients have an increased risk of developing fungal necrotising otitis externa [11, 12]. The diagnosis is often based entirely on the clinical signs and symptoms such as swelling, redness of the skin, itching, detritus, moisture, pain and discharge. Unfortunately, these symptoms are unspecific with the obvious exception of visible fungal growth [8, 13, 14]. Culture and sensitivity (C + S) testing can help secure the diagnosis [15, 16]. A multitude of antifungal drugs, dyes and antiseptics are used [1, 6, 17, 18]. Some (acetic acid and gentian violet) are known to be ototoxic in animal studies, a few are proven safe (clotrimazole, miconazole and nystatin), but the ototoxicity status of many drugs remains unknown [1]. Treatment is mainly topical. In Denmark, the diagnosis and treatment of otomycosis is primarily provided by private ear, nose, and throat (ENT) clinics. The aim of the present study was to investigate how otomycosis is diagnosed and treated by private ENT consultants. We also investigated if the patient’s immune status and the presence of a tympanic membrane perforation affected the chosen treatment.

METHOD

The Danish Healthcare Services’ website [19] identified 147 active private ENT consultants who were sent a two-page anonymised questionnaire electronically. Subsequently, we also sent the questionnaire by surface mail to non-responders (Table 1). The questionnaire was designed so that the answers were mainly yes/no or the name of a treatment in order to facilitate data entry and reduce the time needed to fill in the questionnaire as most private ENT consultants have a tight schedule. No validated questionnaire on the diagnosis and treatment of otomycosis was found in the literature. Therefore, we designed a questionnaire specifically for this study. The Danish Association of Research-interested Otorhinolaryngology Consultants (SAFSOD) tested the questionnaire and adjustments were made according to their comments. No further validation was conducted. Data were entered and analysed in IBM SPSS Statistics 22. Differences between groups were tested by Fisher’s exact test. Random samples were checked for input errors.

Trial registration: No trial registration was necessary as the questionnaire was anonymous and contained no patient data.

RESULTS

The questionnaire response rate was 70%. The number of otomycosis patients seen per year per consultant is shown in Table 2. The diagnosis was generally based on the clinical otomicroscopic presentation as only 20% required to see fungal hyphae. However, it should be noted that yeasts do not produce hyphae [20]. At the primary visit, 42% sent material for C + S testing and immediately started treatment. However, 92% sent material for C + S testing if the primary treatment failed and 95% performed intensive aural cleaning using an otomicroscope. A variety of medical treatments with or without antiseptics, dyes, hydrocortisone and antibacterial drugs were used (Table 3). Treatment was changed if the primary and/or secondary treatment was ineffective. The diversity seen in the initial treatment continued when treatment was changed. The treatment received by those who did not experience treatment failure showed the same diversity as was seen in the other groups.

The results for antifungal drug administration in immunocompetent patients are shown in Table 4. 2% without tympanic membrane perforation (–p) and 5% with tympanic membrane perforation (+p) were treated with systemic antifungal drugs. There was no statistical difference between the two groups (p = 0.45, Fisher’s exact test). The same applied to the use of topical antifungal drugs (90% –p/85% +p; p = 0.84, Fisher’s exact test). Thus, the presence of a tympanic membrane perforation did not alter the drug administration route. The preferred antifungal drugs were miconazole, clotrimazole, ketokonazole and econazolnitrate (topical imidazoles); fluconazole and itroconazole (systemic imidazoles); and nystatin (topical polyene) and terbinafine (topical allylamine).

27% –p /23% +p were treated with antiseptics alone or in combination with other antifungal drugs as the primary treatment. There was no statistically significant difference between the two groups (p = 0.75, Fisher’s exact test). Thus, the presence of a tympanic membrane perforation did not affect the use of antiseptics. Methylrosanilin (gentian violet) was the preferred antiseptic dye. Atamon and Vioform were the second and third preferences, respectively. 27% –p/19% +p used hydrocortisone alone or in combination with other drugs as the initial treatment.

13% reported that they treated immunocompromised patients differently. The main difference was that 77% sent material for C + S at the first visit compared with 23% if the patients were immunocompetent (p = 0.17). The primary line of treatment was still topical antifungal drugs which showed the same diversity as was seen in the immunocompetent patients. However, if the primary treatment failed, treatment was intensified with systemic antifungal drugs, more frequent aural cleaning, conference with a microbiologist and referral to hospital.

DISCUSSION

Treatment of otomycosis can be troublesome and may involve several treatment modalities. It is therefore important to collect systematic knowledge about the diagnostic habits and potential variations concerning treatment in order to improve the quality of the diagnosis as well as the treatment.

In the present study, we found that 80% treated otomycosis based on non-specific clinical signs without visible hyphae. 42% performed C + S testing before starting therapy, while 92% did C + S testing if the initial therapy failed. 95% performed otomicroscopic intensive aural cleaning. 89% used topical drugs, while only 2% used systemic treatment. 80% used antifungal drugs initially, while 5% used antiseptics only. The use of antiseptics increased to 10% if treatment failure was experienced. Perforation of the tympanic membrane did not influence the choice of treatment.

Diagnosis and symptoms

Previous studies have found that clinical symptoms of otomycosis were unspecific and recommended C + S testing unless characteristic fungal hyphae were seen (yeasts do not produce hyphae). Saunders et al [7] made a retrospective study of 170 patients with external otitis and found that fungi were involved in 20% with a mixed bacterial and fungal aetiology, while 5% were positive for fungi alone. Only 38% of positive fungal cultures had clinical signs of fungi. Kurnatowski & Filipiak [8] studied 249 patients with external otitis and found that 15% had a mixed bacterial and fungal aetiology, while 13% were caused by fungus alone. The most frequent symptoms were pain and wet grey debris. In addition, Ho et al [9] and Kaur et al [15] found unspecific symptoms such as pruritus, aural fullness, hearing loss, tinnitus, discharge and pain.

Culture and sensitivity testing

Vennewald & Klemm [1] recommended a sterile swab and debris sample from the ear canal for C + S testing. The reliability of C + S results was questionable if it was only performed when treatment failed. The frequent aural cleaning and topical antifungal treatment before C + S may explain why no causative agent was found.

Antifungal drugs

In our study, 80% of the initial treatment (Table 3) was antifungal drugs alone or in combination with antiseptics and or hydrocortisone. The use of topical imidazoles, mainly clotrimazole and miconazole, were preferred.
Imidazoles have a broad spectrum against yeasts (Candida spp.), molds (Aspergillus spp.), dermatophytes and gram-positive bacteria and are safe in animal studies. However, the ototoxicity in humans has not been tested. The same is true for the polyene nystatin. Terbinafine is an allylamine which primarily works against dermatophytes, but its ototoxicity has not been tested [1]. Some of the treatment failures experienced in our study may result from a relatively frequent use of terbinafine in combination with only 42% performing C + S testing before starting treatment. Patient compliance and a too short period of treatment may also affect treatment efficacy.

Antiseptics

27% used some kind of antiseptic therapy. The preferred antiseptic was methylrosanilin (gentian violet), which is ototoxic in animal models [1], but no studies have been made on humans. This dye was regularly used, and we assume that the ENT consultants did not experience obvious deterioration in the patients’ hearing. The application method of methylrosanilin was most often basting of the external ear canal skin with a cotton stick soaked in methylrosanilin and the method was therefore also considered safe even in patients with tympanic membrane perforation.

Hydrocortisone

Hydrocortisone is known to reduce oedema and the immunological response and may therefore have a place in otomycotic treatment of itching and pain. It may also be beneficial for patients who are susceptible to otomycosis due to an underlying skin disease.

Tympanic membrane perforation

In our study, the presence of a tympanic membrane perforation did not alter the choice of treatment. Hurst [3] studied 22 cases of otomycosis with tympanic membrane perforation. Patients were treated with a combination
of topical clotrimazole, hydrocortisone and antibacterial drugs. All but one patient with a completely disintegrated tympanic membrane recovered without persistent hearing loss. Given that the ototoxicity of the majority of antifungal drugs used remain unknown, the failure to consider a tympanic membrane perforation may not be advisable. Recommendation of frequent cleaning and the use of antifungal drugs that have been proven safe in animal studies may therefore be advisable.

Immune status

In accordance with Rutt & Staloff [13] and Viswanatha et al [10], immunocompromised patients can be treated efficiently and safely with topical antifungal drugs, but vigilance and prompt treatment is recommended. In this study, the 13% who chose to treat immunocompromised patients differently also consulted early on with a microbiologist and were referred to hospital in case of signs of treatment failure.

Strengths and limitations of our study

This study is the first of its kind in Denmark. The response rate was 70% (103 out of 147), which we find acceptable. There are a number of limitations in our study. Among these are recall bias when filling in the questionnaire and selection bias of responders and non-responders. A response rate above 80% would have been preferable, but as this study demonstrates a very large diversity in the used treatments, it seems very unlikely that a higher response rate would have changed this conclusion. Before using the questionnaire, it was tested by a group of private ENT consultants who formed part of the target group.

Future research

No existing guideline or randomised controlled efficacy and outcome studies for the treatment of otomycosis were found in the literature. Future studies may include: diagnostic criteria, when to perform C + S, the optimal way to sample material for C + S testing, interval for aural cleaning, preferred initial treatment (including dosis), length of therapy and choice of therapy in case of treatment failure. Furthermore, recommendations are needed on whether a tympanic membrane perforation or immunosuppression should alter treatment.

CONCLUSION

The diagnosis of otomycosis in private ENT clinics was based on non-specific clinical signs. 42% performed C + S testing at the first visit. Initial treatment was empirical and in case of failure 100% performed C + S. Almost all performed intensive otomicroscopic aural cleaning. A large variety of antifungal drugs, hydrocortisone, antiseptics, dyes and antibacterial drugs were administered topically. Tympanic membrane perforation did not affect choice of treatment. 13% treated immunocompromised patients more vigorously. This study revealed large discrepancies in how otomycosis is diagnosed and treated. Additional research is needed to provide advice and to ensure evidence-based diagnosis and treatment of otomycosis.

Correspondence: Elisabeth Arndal. E-mail: elisabetharndal@hotmail.com

Accepted: 29 February 2016

Conflicts of interest:Disclosure forms provided by the authors are available with the full text of this article at www.danmedj.dk